Speech-to-speech over WebSockets
Realtime API
Low-latency voice conversations over the OpenAI-compatible /v1/realtime WebSocket endpoint.
wss://api.llmgateway.io/v1/realtime
Can you move my 3pm to tomorrow?
Done. It's now tomorrow at 3pm.
gpt-realtime · voice activity detection
01
Low latency
Audio streams both ways.
02
Voice detection
Server-side turn taking.
03
Function calling
Act mid-conversation.
04
Browser safe
Short-lived client secrets.
Built for
Voice agents
Support and booking lines
Live tutoring
Natural back-and-forth
Hands-free apps
Talk instead of type
Open a session
app.ts
1import WebSocket from "ws";2 3const ws = new WebSocket(4 "wss://api.llmgateway.io/v1/realtime?model=gpt-realtime",5 { headers: { Authorization: "Bearer " + process.env.LLM_GATEWAY_API_KEY } },6);7 8ws.on("message", (message) => console.log(JSON.parse(message.toString())));